Nicheloom

Market intelligence for builders — see what's gaining traction before it's crowded.

Oats

Open source meeting notes that never leave your Mac

Details

External ID
1244248
Source
PH
Company
—
Product
Oats
Website domain
producthunt.com
Launched
Sept. 10, 2026
Cohort
—
Upvotes
1
Upvotes percentile
0.30815693820825313
Tags
Productivity, Open Source, Meetings, GitHub
Fetched at
Sept. 10, 2026, 5:15 p.m.
Updated at
Sept. 10, 2026, 5:15 p.m.

Description

Oats records, transcribes, and summarizes your meetings entirely on your Mac. No bot joins your call, no audio leaves your machine, no account, no subscription. Live transcription with speaker labels, action items collected across meetings, a knowledge graph linking people, projects, and topics, and a Spotlight-style Ask popup. Pick your models: Apple Intelligence, Ollama, Whisper, or Parakeet. Notes are plain files on disk. Free and open source under Apache 2.0.

Enrichment

Theme
AI voice recorders and meeting notetakers
Vertical
Horizontal
Function
Communication
Audience
B2C
AI stance
Not AI
Project type
Hobby / open-source project
Normalized one-liner
meeting notes app for mac
Manually corrected
False

Could you build this?

Partial The UI and local DB are vibe-codeable, but robust on-device meeting transcription requires system-level audio capture (loopback drivers) and local ML model orchestration.

What it would actually take: Requires a native macOS app (Swift/Objective-C) utilizing ScreenCaptureKit or virtual CoreAudio HAL drivers to capture multi-app audio without meeting bots. It requires bundling and optimizing quantized models like Whisper.cpp or MLX for local real-time ASR, along with speaker diarization models (e.g. pyannote converted to CoreML) and local LLM execution (llama.cpp) for graph summarization without thermal throttling.

Competitors

Other products that read as similar to this one — 174 launches clear the similarity bar, closest 8 shown.

Attention rank: #107 of 175 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).

Launched 279 days after the earliest competitor.

Other launches for this product